IndexFlow publishes search indexing tools as self-hosted Rust crates

IndexFlow is open-source infrastructure for the technical part of search indexing, written in Rust and published as crates. Created on 4 August 2026, 59 stars at the time of writing, licensed MIT or Apache 2.0.

What it covers

Search engine discovery. Sitemap processing. Technical validation. URL indexing workflows. Submission to search engines. Multi-site scheduling.

Two crates are published: indexflow-seo and indexflow-sitemap. The repository description names three specific features: inline quality gates during indexing, fair scheduling across multiple sites, and a rolling quota circuit breaker.

It is self-hostable, and there is a hosted service behind it at indexflowing.com, which is the usual arrangement of free open-source code with a paid hosted version.

The three features are the reason to look

Everything IndexFlow does can be done with scripts, and in most organisations it is. The scripts are the problem.

A rolling quota circuit breaker matters because submission APIs have limits, and the failure when you exceed one is not always visible. A script that silently gets rejected for a week looks identical to a script that is working. A circuit breaker that tracks a rolling window and stops before the limit is the difference between a system and a cron job.

Fair multi-site scheduling matters if you run more than one website. The simple version processes sites in order, so the last one gets whatever quota is left, which is usually none.

Inline quality gates mean the checks run during the indexing workflow rather than as a separate audit nobody runs.

None of that is difficult. All of it is the kind of thing a script gains after three incidents, and having it as a library with a version number is better than having it as a file called submit.sh that one person understands.

What to check first

The licence field on the repository is empty in the GitHub API even though the README states MIT or Apache 2.0. That is almost certainly a missing LICENSE file rather than a real ambiguity, and it is worth confirming before you depend on it.

At 59 stars and under a month old, this is early. Rust and a small number of dependencies help, since a library that reads sitemaps and calls an HTTP endpoint can do little damage even if it is wrong.

The specific case where this is worth ten minutes is if you already submit URLs to IndexNow or Google Search Console from a script you wrote yourself, and you do not currently know what happens when that script reaches a quota. Read the circuit breaker code, and either use it or copy the idea.

Documentation is on docs.rs, which is the right place for it.

Source

• IndexFlowing/IndexFlow-core on GitHub. Star and date figures read from the GitHub API on 31 August 2026